Phosphoinositide 30H-kinases (PI3Ks) are a large family of enzymes capable of 3-phosphorylating at least one of the cellular phosphoinositides (Whitman et al., 1988, Nature 332:644-646; Auger et al., 1989, Cell 57:167-175). 3-phosphorylated phosphoinositides are found in all higher eukaryotic cells. A growing body of evidence implicates PI3K and a lipid product of this enzyme, phosphatidylinositol (3,4,5) -triphosphate (hereinafter "PtdIns (3,4,5)P.sub.3 "), as part of a novel and important second messenger system in cellular signal transduction. The components of this novel PtdIns(3,4,5)P.sub.3 -based signalling system appear to be independent of the previously characterized signalling pathway based on inositol phospholipids, in which a phosphoinositidase C (PIC) hydrolyses PtdIns(4,5)P.sub.2 to release the structurally distinct second messengers inositol (1,4,5)-triphosphate (Ins(1,4,5)P.sub.3) and diacylglycerol.
Select extracellular agonists and growth factors will stimulate intracellular PI3K activity and cause the rapid and transient intracellular accumulation of PtdIns(3,4,5)P.sub.3. Surprisingly, stimulation of a variety of different types of cell surface receptors, including receptor tyrosine kinases, receptors associated with src family non-receptor tyrosine kinases, cytokine growth factors, and G protein coupled receptors will all activate members of the P13K family. (Reviewed in Stephens et al., 1993, Biochemica et Biophysica Acta, 1179:27-75). For example, tyrosine kinase receptors which, when activated, result in increased accumulation of PtdIns(3,4,5) P.sub.3 are the PDGF receptor, the EGF receptor, members of the FGF receptor family, the CSF-1 receptor, the insulin receptor, the IGF-1 receptor, and the NGF receptor. Receptors associated with src family non-receptor tyrosine kinases which stimulate PtdIns(3,4,5)P.sub.3 accumulation are the Il-2 receptor, Il-3 receptor, mIgM receptor, the CD4 receptor, the CD2 receptor, and the CD3/T cell receptor. Additionally, the cytokine Il-4 receptor and the G protein linked thrombin receptor, ATP receptor, and the fMLP receptor all stimulate the activity of a PI3K, resulting in subsequent PtdIns(3,4,5)P.sub.3 accumulation. Thus, PtdIns(3,4,5)P.sub.3 appears to be a second messenger in extremely diverse signalling pathways.
Support for the proposition that PI3K activity and production of PtdIns(3,4,5)P.sub.3 is a physiological relevant pathway of signal transduction for these diverse receptors is derived, inter alia, from two different lines of experimental evidence: inhibition of PI3K activity by fungal metabolites and observations of direct protein associations. Wortmannin, a fungal metabolite, irreversibly inhibits PI3K activity by binding covalently to the catalytic domain of this enzyme. Inhibition of PI3K activity by wortmannin eliminates the subsequent cellular response to the extracellular factor. For example, neutrophils respond to the chemokine fMet-Leu-Phe (fMLP) by stimulating PI3K and synthesizing PtdIns(3,4,5)P.sub.3. The synthesis correlates with activation of the respiratory burst involved in neutrophil destruction of invading microorganisms. Treatment of neutrophils with wortmannin prevents the fMLP-induced respiratory burst response. Thelen et al., 1994, PNAS, USA 91:4960-4964. Indeed, these experiments with wortmannin, as well as other experimental evidence, shows that PI3K activity in cells of hematopoietic lineage, particularly neutrophils, monocytes, and other types of leukocytes, is involved in many of the non-memory immune responses associated with acute and chronic inflammation.
PI3K enzymes interact directly with, and may be co-purified with, activated forms of several receptor tyrosine kinases. When purified, receptor tyrosine kinase associated PI3K was found to consist of 170-200 kD heterodimers (Otsu et al., 1991, Cell 65:91-104, Pons et al., 1995, Mol. Cell. Biol. 15:4453-4465, Inukai et al., 1996, J. Biol. Chem. 271:5317-5320) comprising a catalytic subunit and an adaptor (or regulatory) subunit.
Two different homologs of the catalytic subunit, 110 .alpha. and 110.beta., have been described and cloned. The catalytic subunit, which irreversibly binds wortmannin, tightly associates with one or other members of a small family of highly related regulatory subunits, p55.alpha., p55P1K, p85.alpha. and p85.beta., to form the 170-200 kD heterodimers. The known regulatory subunits contain a large collection of protein:protein interaction domains, including two SH2 domains (Cantley et al., 1991, Cell 64:281-302).
The presence of the SH2 domains are thought to be responsible for the binding and stimulation of PI3K heterodimers to activated receptor tyrosine kinases. Activated receptors are phosphorylated at key tyrosine residues within local consensus sequences preferred by the SH2 domains found in the 55-87 kD PI3K adaptors (Songyang et al., 1993, Cell 72:767-778). Once the PI3K heterodimer binds, it directly activates the PI3K catalytic subunit (although this effect is relatively small in vitro, Carpenter et al., 1993, J. Biol. Chem. 268:9478-9483, Backer et al., 1992, EMBO J. 11:3469-3479) and translocates the cytosolic PI3K to a source of its phospholipid substrate. The combination of these factors leads to a surge in PtdIns(3,4,5)P.sub.3 production. Clearly, these isoforms of PI3Ks (p100.alpha./110.beta./p55.alpha.,p55PIK) seem structurally adapted to function as dedicated signal transducers downstream of receptor-regulated tyrosine kinases, very like the way the .tau.-family of PI-PLCs are regulated by receptor-sensitive tyrosine kinases (Lee and Rhee, 1995, Current Biol. 7:193-189).
However, the 110/p85 sub-family of PI3Ks do not seem to be involved in the production of PtdIns(3,4,5)P.sub.3 that can occur as a result of activation of cell surface receptors which utilize heterotrimeric GTPases to transduce their signals (e.q., fMLP, PAF, ATP, and thrombin). These types of cell surface receptors have been primarily described in cells of hematopoietic origin whose activation is involved inflammatory responses of the immune system. Recent evidence has suggested that a chromatographically distinct form of wortmannin-sensitive PI3K is present in U937 cells and neutrophils that possesses a native, relative molecular mass of about 220 kD (Stephens et al., 1994, Cell 77:83-93). This PI3K activity can be specifically stimulated by G.beta..gamma. subunits, but not G.alpha.-GTP subunits. A similar PI3K activity has also been described in an Osteosarcoma cell line (Morris et al., 1995, Mol. Pharm. 48:532-539). Platelets also contain a G.beta..gamma.-sensitive PI3K, although it is unclear whether this is a p85/110 PI3K family member (Thomason et al., 1994, J. Biol. Chem. 269:16525-16528). It seemed likely that this poorly characterized, G.beta..gamma.-sensitive PI3K might be responsible for production of PtdIns(3,4,5)P.sub.3 in response to agonists like ATP, fMLP etc.
Stoyanov et al., (1995) have recently published the cloning and expression of a wortmannin-sensitive PtdIns(4,5)P.sub.2 -selective PI3K, termed 110.gamma., from a human bone marrow cDNA library. 110.gamma. was amplified by PCR using primers designed to target potential PI3K's as well as PtdIns4-kinase's catalytic centers. It is clearly distinct from 110.alpha. and 110.beta., as it lacks, for example, an amino-terminus binding domain for a member of the p85 adaptor family. 110.gamma. was speculated to be the PI3K activity downstream of heterotrimeric GTPase-linked receptors on the basis of its sensitivity to both G.alpha.-GTP and G.beta..gamma.-subunits in vitro and its expression in myeloid-derived cells. Nevertheless, this hypothesis left several unresolved questions regarding the earlier biochemical evidence which indicated that the G.beta..gamma. responsive PI3K was not stimulated by G.alpha.-GTP subunits, and that it possessed a much greater molecular mass of about 220 kD.
The effects of G.beta..gamma. subunits on 110.gamma. were suggested to be mediated via a putative NH.sub.2 -terminus pleckstrin homology (PH) domain. However, with the description of an increasing number of G.beta..gamma. regulated effectors, mounting evidence suggests that PH domains do not represent a widely used G.beta..gamma. binding domain. Recent work, using a panel of relatively small peptides based on the sequence of domains only found in the G.beta..gamma.-activated adenylate cyclases (ACs 2 and 4) which specifically block G.beta..gamma. activation or inhibition of several effectors, has suggested there may be some grounds for believing G.beta..gamma. subunits contain a widely used effector activating domain. Further, regions in different effectors that interact with this effector activating domain show significant sequence similarities. Hence a motif (Gln-X-X-Glu-Arg) within the domain in AC2 highlighted by these peptide studies also appears in regions of potassium channels and .beta.-ARKs already implicated in regulation by G.beta..gamma. subunits (Chen et al., 1995, Science 268:1166-1169). However, this motif is not replicated in all proteins known to be regulated by G.beta..gamma. subunits, and consequently sequence analysis cannot currently predict whether a protein will be regulated by G.beta..gamma. subunits.
Identification of the mechanism by which PI3K activity is activated by cellular agonists which transduce their signals through G protein linked receptors is lacking. It is important to note that the vast majority of agonists which activate the neutrophil respiratory burst involved in the inflammatory response will bind to G-protein-coupled receptors rather than receptor tyrosine kinases. Thus, the mechanism by which PI3K is regulated in response to these types of chemokines is likely to be very different from regulation by growth factors which signal through tyrosine kinases. The present invention is directed towards resolving this issue by the identification, purification, and cloning of a novel and specific form of PI3K which is activated by .beta..gamma. subunits of trimeric G-proteins.